Sarkar Souvik, Alurwar Aniket, Ly Carole, Piao Cindy, Donde Rajiv, Wang Christopher J, Meyers Frederick J
Divisions of Gastroenterology, Hepatology and Hematology/Oncology, Department of Internal Medicine, University of California, Davis, Sacramento, California.
Center for Precision Medicine and Data Sciences, University of California, Davis, Sacramento, California.
Gastro Hep Adv. 2024 Jan 23;3(4):498-505. doi: 10.1016/j.gastha.2024.01.007. eCollection 2024.
Hepatocellular carcinoma (HCC) incidence is increasing and correlated with metabolic dysfunction-associated steatotic liver disease (MASLD; formerly nonalcoholic fatty liver disease), even in patients without advanced liver fibrosis who are more likely to be diagnosed with advanced disease stages and shorter survival time, and less likely to receive a liver transplant. Machine learning (ML) tools can characterize large datasets and help develop predictive models that can calculate individual HCC risk and guide selective screening and risk mitigation strategies.
Tableau and KNIME Analytics were used for descriptive analytics and ML tasks. ML models were developed using standard laboratory and clinical parameters. Sci-kit learn algorithms were used for model development. Data from University of California (UC), Davis, were used to develop and train a pilot predictive model, which was subsequently validated in an independent dataset from UC San Francisco. MASLD and HCC patients were identified by International Classification of Diseases-9/10 codes.
Of the patients diagnosed with MASLD (n = 1561 training; n = 686 validation), HCC developed in 14% (n = 227) of the UC Davis training cohort and 25% (n = 176) of the UC San Francisco validation cohort. Liver fibrosis determined by the noninvasive Fibrosis-4 score was the strongest single predictor for HCC in the model. Using the validation cohort, the model predicted HCC development at 92.06% accuracy with an area under the curve of 0.97, F1-score of 0.84, 98.34% specificity, and 74.41% sensitivity.
ML models can aid physicians in providing early HCC risk assessment in patients with MASLD. Further validation will translate to cost-effective, personalized care of at-risk patients.
肝细胞癌（HCC）的发病率正在上升，且与代谢功能障碍相关脂肪性肝病（MASLD，原非酒精性脂肪性肝病）相关，即使在没有晚期肝纤维化的患者中也是如此，这些患者更有可能被诊断为晚期疾病阶段且生存时间较短，接受肝移植的可能性也较小。机器学习（ML）工具可以对大型数据集进行特征描述，并有助于开发预测模型，该模型可以计算个体HCC风险，并指导选择性筛查和风险缓解策略。
使用Tableau和KNIME Analytics进行描述性分析和ML任务。使用标准实验室和临床参数开发ML模型。使用Sci-kit learn算法进行模型开发。来自加利福尼亚大学（UC）戴维斯分校的数据用于开发和训练一个试点预测模型，随后在来自UC旧金山分校的独立数据集中进行验证。通过国际疾病分类-9/10编码识别MASLD和HCC患者。
在诊断为MASLD的患者中（n = 1561例用于训练；n = 686例用于验证），UC戴维斯分校训练队列中的14%（n = 227）和UC旧金山分校验证队列中的25%（n = 176）发生了HCC。由无创性Fibrosis-4评分确定的肝纤维化是模型中HCC最强的单一预测因素。使用验证队列，该模型预测HCC发生的准确率为92.06%，曲线下面积为0.97，F1分数为0.84，特异性为98.34%，敏感性为74.41%。
ML模型可以帮助医生对MASLD患者进行早期HCC风险评估。进一步的验证将转化为对高危患者具有成本效益的个性化护理。
